Artificial Intelligence in Manufacturing Quality Assurance: Applications, Challenges, and Opportunities for Emerging Industrial Economies

Abstract

Quality assurance (QA) in manufacturing has long depended on manual inspection, sampling, and rule-based checks, which are slow, vary between inspectors, and cover only part of production. This paper is a narrative review of how artificial intelligence (AI) is used in manufacturing QA. It covers deep-learning visual inspection, predictive quality models built on process data, anomaly detection, and predictive maintenance. It summarizes the benefits reported in the literature, the main limitations (labelled data, changing conditions, explainability, and computing cost), and the ISO 9001 context. It then considers what these findings mean for manufacturers in developing countries and in Rwanda, using national accounts data (manufacturing was about 8% of Rwanda's GDP in 2024 and grew 10% in 2025) and the Made in Rwanda policy, which has made product quality a national priority. The paper argues that AI should support QA professionals rather than replace them, and proposes an eight-step adoption roadmap built around small, well-defined pilot projects. It reports no new experimental data.